and sorry for this question, but it's been on my mind lately.

I'll start by telling you that I have never worked with a calibrator in my life. But I THINK I know how it should work.

I know that (at least in my country) to calibrate an instrument you need to use a calibrator 10x more precise than your instrument's precision (if you have a voltage/current meter with +-1%, you need a calibrator with +-0,1%). The instrument is adjusted according to the calibrator's output.

If this is how a calibrator works, then what's the difference between a calibrator and a precision PSU? Is the difference that the calibrator has a wider output range and/or formats? Is there anywhere that I can find the block diagram specification of a calibrator (like this that you can find on TI's page)?

Are there any other functionalities that I'm missing?

PS: I don't have a calibrator that I can crack open and buying an old/cheap calibrator from ebay is not an option (ebay doesn't send to my country and even if they did the taxes on it would bankrupt me.)

  • \$\begingroup\$ Calibration is a process involving multiple points in time, so you do not only need an instrument being precise at one point in time, but not drifting over a long period of time. \$\endgroup\$
    – PlasmaHH
    Commented Dec 10, 2015 at 21:03
  • \$\begingroup\$ @PlasmaHH So the only difference between a calibrator and a precision power supply is that the calibrator doesn't drift over time? I imagine that's kinda difficult to accomplish .. don't know .. I'm certainly going to study. This seems really interesting \$\endgroup\$
    – morcillo
    Commented Dec 10, 2015 at 21:08
  • \$\begingroup\$ @PlasmaHH From what I can see drift is the variation of the output (or sense circuit) due to temperarture, materials, etc (environmental changes to the circuit that cause it to lose precision). So ... this means that a calibrator is differente because it has precision components and in some way tries to account for its environment (temperature control, cabling, size, weight, etc) so that the output will always be reliable? Am I right or just delusional? \$\endgroup\$
    – morcillo
    Commented Dec 10, 2015 at 21:17
  • 1
    \$\begingroup\$ A voltage calibrator is designed with special attention to maintaining a constant specified output voltage for a reasonable variation of environmental conditions. It would probably produce one, or very few, voltages, but unlike a power supply would not be intended to produce any significant current. \$\endgroup\$ Commented Dec 10, 2015 at 21:56
  • 1
    \$\begingroup\$ "cuntionalities" how rude! But on a more sensible theme, what are you trying to achieve? \$\endgroup\$
    – Andy aka
    Commented Dec 11, 2015 at 0:26

2 Answers 2


As stated above, a "Calibrator" is indeed a device used to calibrate another device. There are many types though.

(The below only deals with Voltage as per the original question)

A Primary Standard is a standard based on absolute, immutable physical properties of a system. For voltage, they typical primary standard is a Josephson Junction Array.

A "Working" or "Secondary" standard is typically a high precision voltmeter, with very low drift over time or temperature characteristics. Even so, they are designed to work only over a limited temperature range. Working standards are usually guaranteed to maintain their calibration over a certain period of time, anywhere from one week to several years. Instrument manufacturers generally target 1 year. Working standards need to be calibrated on a periodic basis to either a higher precision working standard, or preferably a primary standard.

In the USA, a Working standard is typically calibrated on a periodic basis to an NIST (National Institute of Standards and Technology) traceable source. That means the calibration can actually be traced back to a certified primary standard.

The Working Standard is what an instrumentation tech would typically use to calibrate an actual field device.

Just a note, that all 'Field Calibration' references need to be calibrated or at least checked against a primary source. Even something as simple as weights used for calibrating scales. Over time, the actual weight can change just from absorption of atmospheric gasses. When a lab checks a purely physical standard, they don't generally modify it to make it equal to the reference. For a mass (weight), they will not typically add or remove mass, rather they will issue a certificate that states what the actual mass of the object is.


I would not call the "thing that make a precise voltage" a calibrator - I'd call it a reference, specifically a voltage reference.

My voltage reference is a "standard cell battery" which has not, itself, been compared to another reference for many a year, so it's not "officially" good for much, but for my own purposes it's fine, and does not incur the cost of shipping it off to a standards laboratory for comparison to another standard cell (or several) and application of a sticker stating that it's calibrated, typically on an annual basis.

The design of such a cell is far more about providing a very specific voltage, and not much at all otherwise acting like a "battery" - it would be a terrible source of power. Since it's very likely a Weston Cell it would also send the RoHS folks running for the hills in terror, with both cadmium and mercury being involved.

There are other types of voltage reference, some of which might scare the RoHS folks less. Evidently the "primary standard" (against which lesser, or secondary standards are judged and the "offical volt" defined) is now a low-temperature superconducting standard, based on work that won Brian Josephson the Nobel Prize in Physics in 1973. Presumably it's mercury and cadmium free.

Likewise a current reference, a standard resistor (or a set of standard resistors) etc...


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.